Synthesis and oxidation of 'non-annulated' vitamin E-like compounds

Article abstract of DOI:10.1016/j.tetlet.2005.09.009

Two model compounds (3 and 4) having the typical α-tocopherol-type substitution pattern of the aromatic ring, but lacking the annulated pyran ring, have been synthesized. Upon oxidation, the two possible ortho-quinone methides (oQMs) of each are formed in

Full text of DOI:10.1016/j.tetlet.2005.09.009

Tetrahedron
Letters
Tetrahedron Letters 46 (2005) 7845–7848
Synthesis and oxidation of ‘non-annulated’ vitamin E-like
compounds
Thomas Rosenaua,* and Amnon Stangerb
aDepartment of Chemistry, University of Natural Resources and Applied Life Sciences, Vienna (BOKU),
Muthgasse 18, A-1190 Vienna, Austria
bDepartment of Chemistry, Technion—Israel Institute of Technology, The Institute of Catalysis Science and Technology and
The Lise–Meitner–Minerva Center for Computational Quantum Chemistry, Haifa 32000, Israel
Received 9 May 2005; revised 26 August 2005; accepted 2 September 2005
Available online 22 September 2005
Abstract—Two modelcompounds ( 3 and 4) having the typical a-tocopherol-type substitution pattern of the aromatic ring, but lack-
ing the annulated pyran ring, have been synthesized. Upon oxidation, the two possible ortho-quinone methides (oQMs) of each are
formed in equal ratio. DFT calculations suggest that there is no angular strain in 3 and 4, and each of the oQM pairs is of similar
energy. These results prove that the commonly observed regioselectivity in oxidations of vitamin E-type model compounds is not an
electronic substitution effect, but a consequence of annulation.
ꢀ 2005 Elsevier Ltd. All rights reserved.
1.Introduction
selectivity is instead a function of angular strain, chang-
ing gradually with the sum of the two annulation angles,
but not abruptly with the size of the annulated ring. As a
consequence, the oxidation regioselectivity of vitamin
E-type phenols was fully explained by the theory of
strain-induced bond localization (SIBL).5
Phenolic compounds derived from trimethylhydroquin-
one by annulation are commonly referred to as vitamin
E-type antioxidants. They possess the typicalsubstitu-
tion pattern of a-tocopherol( 1),1 that is, three aromatic
methylgroups and a fourth substituent, which forms a
ring structure with one of the oxygens. Oxidation of
these compounds to ortho-quinone methides2 (oQM)
proceeds with peculiar regioselectivity, apparently exclu-
sively involving C-5a (but not C-7a). In the pertinent lit-
erature, this behavior has been commonly explained by
the so-called Mills–Nixon effect.3 From this, it was pre-
dicted that oQM formation proceeds exclusively involv-
ing the ÔupperÕ C-5a methylgroup in benzopyranosl,
such as 1 and 2, and exclusively involving the ÔdownÕ
Another older explanation attempt—mostly evoked
together with the Mills–Nixon effect and still present
in todayÕs literature—has been postulating the electronic
effects of the annulated six-membered ring to be the cru-
cialfactors in governing the oxidation regioselectivity of
vitamin E-type antioxidants. Even though it seems diffi-
cult to imagine why the small difference of one CH2
group—with regard to inductive effects—should be able
to change the regioselectivity completely when going
4
C-7a methylgroup in benzofuranosl. In a recent study,
5a
we have shown that the Mills–Nixon theory from 1930
was based on the wrong premises, even though it seemed
to provide the proper explanations for the vitamin E sys-
tem. It was moreover shown that the prediction of com-
plete regioselectivity in oxidations of vitamin E-type
antioxidants was contradictory to experiment: the regio-
4
5
6
7
4a
HO
7a
3
1 R = C16H33
2 R = CH3
2a
R
1
O
2
8a
8
8b
3
C16H33
=
6
3
4
HO
HO
4
Keywords: Tocopherol; Vitamin E; Antioxidants; Mills–Nixon effect;
Strain-induced bond localization; SIBL.
Corresponding author. Tel.: +43 1 36006 6071; fax: +43 1 36006
6059; e-mail: thomas.rosenau@boku.ac.at
2
2
5
5
1
1
O
O
*
6
3
4
0040-4039/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2005.09.009

7846
T. Rosenau, A. Stanger / Tetrahedron Letters 46 (2005) 7845–7848
from six-membered to five-membered annulated rings,
this explanation is still frequently used today.6 It was
the goalof the present study to test the vaildity of this
hypothesis.
bond angles (see Schemes 1 and 2) are 239.3ꢁ and
237.6ꢁ for 3 and 4, respectively. Since these two systems
are not isomeric, it is impossible to compare directly the
heats of formation. However, the results of Eqs. 1 and 2
suggest that there is no special stability or instability
comparing 3 and 4.
2.Results and discussion
3 þ C2H6 ! 4 þ CH4
ð1Þ
DH = 3.1
The common vitamin E modelcompound 2,2,5,7,8-pen-
tamethylchroman-6-ol (2) is oxidized in high (but not
complete) regioselectivity. At room temperature, 98%
of the 5a-oQM was found besides 2% of the 7a-oQM
counterpart. The electronic influences of the annulated
ring consist of inductive effects of the ring carbons on
both C-4a and O-1. Assuming that these effects indeed
govern the observed regioselectivity in oxidations, then
any compound having similar alkyl substituents at both
C-4a and O-1 should exhibit a comparable oxidation reg-
ioselectivity—independent of whether these substituents
are joined in a ring structure or not. Thus, we have syn-
thesized compounds 3 and 4 as two Ôopen ringÕ versions
of 2, which have similar substituents at C-4a and O-1,
but no annulated ring. If indeed the electronic effects
are responsible for the regioselectivity in oxidations of
vitamin E derivatives, then both 3 and 4 should be oxi-
dized in high regioselectivity to the oQM which involves
the 3-methylgroup, whiel the competitive oQM involv-
ing the 5-methyl group should be largely disfavored.
DE(ZPE
corrected) = 2.5 kcalmol ꢀ1
,
kcalmol ꢀ1
3 þ ðCH3Þ2CHCH2CH3 þ CH3OH !
4 þ C2H6 þ ðCH3Þ2CHOH
ð2Þ
DH = ꢀ3.7
DE(ZPE corrected) = ꢀ3.8 kcalmol ꢀ1
,
kcalmol ꢀ1
.
BnO
BnO
i
O
OH
2a
BnO
HO
iii
2
1
ii
O
O
1a
4
Scheme 2. Synthesis of Ônon-annulatedÕ vitamin E modelcompound 4.
Reagents and conditions: (i) H2SO4, toluene, reflux, 10 min; (ii)
MeOTf, 2 h, rt; (iii) H2, Pd/C, 24 h, 33% overall.
Compound 3 was obtained starting from 1-(5-acetoxy-2-
hydroxy-3,4,6-trimethylphenyl)ethanone, which was
readily obtained from TMHQ by treatment with BF3–
acetic acid complex. Claisen etherification of the free phe-
nolic hydroxyl group followed by reduction of the keto
group with simultaneous deacetylation afforded 3 in
73% yield7 relative to the starting ethanone (Scheme 1).
The synthesis of 4 in 33% overall yield8 employed O-
benzylated 2, which underwent ring-opening by an intra-
molecular elimination process, followed by methylation
of the liberated phenolic hydroxyl and simultaneous
debenzylation and hydrogenation of the double bond.
Ag2O, r.t.
CH2Cl2
O
O
3
+
O
O
O
3b
3a
O
EtO
O
EtO
O
In order to obtain some more insight, 3 and 4 underwent
full geometry optimizations at the B3LYP/6-311G* hy-
brid density functionalel vel. 9 Since there are no annula-
tions in these systems, the bond angles should be
relaxed, that is, close to 120ꢁ at the aromatic ring. In-
deed, the sum of the C-2a–C-2–C-1 and O-1a–C-1–C-2
+
O
O
6
5
Ag2O, r.t.
CH2Cl2
O
O
4
+
O
O
O
O
O
O
4a
4b
AcO
AcO
i
O
OH
EtO
O
2a
EtO
O
HO
ii
2
1
+
O
O
O
1a
7
3
8
Scheme 1. Synthesis of Ônon-annulatedÕ vitamin E modelcompound 3.
Reagents and conditions: (i) Me2CHBr, K2CO3, KI, acetone, rt, 24 h;
(ii) Zn, HCl, 2 h, rt, 72%, overall.
Scheme 3. Oxidation of compounds 3 and 4—as the Ônon-annulatedÕ
versions of vitamin E modelcompound 2—to ortho-quinone methides
and trapping in a hetero-Diels–Alder reaction with ethyl vinyl ether.

T. Rosenau, A. Stanger / Tetrahedron Letters 46 (2005) 7845–7848
7847
Oxidation of both compounds 3 and 4 was carried out
Acknowledgements
by either Ag2O or elemental bromine at room tempera-
ture. For determination of the transient ortho-quinone
The financialsupport by the Austrian Fonds zur Fo¨rde-
rung der wissenschaftlichen Forschung, Project P-17428
is gratefully acknowledged. The authors would like to
thank Dr. Andreas Hofinger, Department of Chemistry
at the University of NaturalResources and Applied Life
Sciences (BOKU), Vienna, for recording the NMR
spectra.
methides, we used the fast trapping reaction in excess
10
ethylvinylether, as described in eariler work.
Upon
oxidation, compound 3 afforded the two oQMs 3a and
3b which were trapped as 5 and 6 in a 48/52 ratio inde-
pendent of the oxidant used, which proved that there
was no oxidation regioselectivity at all.11 In the case of
compound 4, the two ortho-quinone methides 4a and
4b were even formed in a perfect 50/50 ratio as reflected
by the 50/50 ratio found for the two trapping products 7
and 8 (Scheme 3).
References and notes
1. Packer, L.; Fuchs, J. Vitamin E in Health and Disease;
MarcelDekker: New York, 1993; For a generalreview on
chromans and tocopherols, see: Parkhurst, R. M.; Skin-
ner, W. A. In Chromans and Tocopherols in Chemistry of
Heterocyclic Compounds; Ellis, G. P., Lockhardt, I. M.,
Eds.; Wiley: New York, 1981; Vol. 36.
2. For a recent review on oQM chemistry, see: van de Water,
R. W.; Pettus, T. R. R. Tetrahedron 2002, 58, 5367–
5405.
3. Original work: (a) Mills, W. H.; Nixon, I. G. J. Chem. Soc.
1930, 2510–2525; See also a later review: (b) Badger, G. M.
Q. Rev. Chem. Soc. 1951, 5, 147.
4. Rosenau, T.; Ebner, G.; Stanger, A.; Perl, S.; Nuri, L.
Chem. Eur. J. 2005, 11, 280–287.
5. For the definition of SIBL (Strain Induced Bond Local-
ization), see: (a) Stanger, A.; Ashkenazi, N.; Boese, R.;
Stellberg, P. J. Organomet. Chem. 1997, 542, 19; Stanger,
A.; Ashkenazi, N.; Boese, R.; Stellberg, P. J. Organomet.
Chem. 1997, 548, 113; Stanger, A.; Ashkenazi, N.; Boese,
R.; Stellberg, P. J. Organomet. Chem. 1998, 556, 249; (b)
Stanger, A.; Tkachenko, E. J. Comp. Chem. 2001, 22,
1377, footnote 6; For recent reviews and investigation of
SIBL/Mills-Nixon effect see also: (c) Frank, N. L.; Siegel,
J. S. In Advances in Theoretically Interesting Molecules;
JAI Press: Greenwich, CT, 1995; Vol. 3, pp 209–260; (d)
It was shown that the oQM, which is formed upon oxi-
dation of the phenol, is similar to the intermediate which
is formed after the rate determining step, which, in turn,
contains the rate determining transition state.4 Thus, the
two possible oQMs for 3 and 4 (3a/3b and 4a/4b, respec-
tively) were calculated. The energy differences between
3a and 3b and 4a and 4b are small: 3b is by
0.1 kcalmol ꢀ1 more stable than 3a (ZPE corrected DE,
both have the same DH), and 4b is by 0.4 kcalmol ꢀ1
more stable than 4a (both ZPE corrected DE and DH),
that is, within the limitations of the theoretical method
each pair shows a comparable stability. The equal stabil-
ity of the different oQM isomers is perhaps best demon-
strated by comparing the energy of the isodesmic
equations 3a with 3b, and 4a with 4b. Within
0.2 kcalmol ꢀ1, each pair of equations shows the same
energy. Thus, it is clear from the calculations that in
the non-annulated systems there is no preference for
one of the two possible oQMs, in strong contrast to
the situation in the annulated systems.
3 þ 4a ! 4 þ 3a
ð3aÞ
DH = ꢀ0.9
´
´
´
´
Maksic, Z. B.; Eckert-Maksic, M.; Mo, O.; Yanez, M.
˜
DE(ZPE corrected) = ꢀ0.9 kcalmol ꢀ1
,
,
,
PaulingÕs Legacy: Modern Modelling of the Chemical
Bond. In Theoretical Computer Chemistry; Elsevier:
Amsterdam, The Netherlands, 1999; Vol. 6, p 47; (e)
Stanger, A.; Vollhardt, K. P. C. J. Org. Chem. 1988, 53,
4889; (f) Rappoport, Z.; Kobayashi, S.; Stanger, A.;
Boese, R. J. Org. Chem. 1999, 64, 4370.
kcalmol ꢀ1
3 þ 4a ! 4 þ 3b
DE(ZPE corrected) = ꢀ0.7 kcalmol ꢀ1
ð3bÞ
DH = ꢀ0.8
kcalmol ꢀ1
6. (a) Isler, O.; Brubacher, G. Vitamins I; George Thieme:
Stuttgart, 1982; p 126; (b) Kamal-Eldin, A.; Appelqvist, L.
A. Lipids 1996, 31, 671–701; (c) Skinner, W. A. J. Med.
Chem. 1967, 10, 657–661; (d) Ref. 3b.
3 þ 4b ! 4 þ 3b
DE(ZPE corrected) = ꢀ0.3 kcalmol ꢀ1
ð4aÞ
DH = ꢀ0.4
7. 3-Ethyl-4-isopropoxy-2,5,6-trimethylphenol (3), fluffy pre-
kcalmol ꢀ1
1
cipitate of white, long needles, mp 123–125 ꢁC. H NMR:
3
d 1.21 (t, 3H, CH2–CH3, J = 7.6 Hz), 1.28 (d, 6H, CH–
3 þ 4b ! 4 þ 3a
ð4bÞ
(CH3)3, 3J = 6.2 Hz), 2.03, 2.09, 2.14 (3 · s, 3 · 3H,
3 · CH3), 2.61 (q, 2H, CH2–CH3, 3J = 7.6 Hz), 4.25 (sept,
1H, CH–(CH3)3, 3J = 6.2 Hz), 4.55 (s, 1H, OH). 13C
NMR: d 11.8, 12.1, 12.9 (3 · CH3), 13.2 (CH2–CH3), 19.8
(CH2–CH3), 22.0 (CH–(CH3)3), 75.7 (CH–(CH3)3), 114.8,
118.3, 119.1, 120.1, 145.6, 146.9 (ArC). Anal. Calcd for
C14H22O2 (222.33): C, 75.63; H, 9.97. Found: C, 75.47; H,
9.86.
DE(ZPE corrected) = ꢀ0.5 kcalmol ꢀ1
,
DH = ꢀ0.4
kcalmol ꢀ1
.
These results clearly disprove the notion that electronic
substituent effects exerted on the aromatic ring via C-4a
and O-1 determine the regioselectivity in oxidations of
vitamin E-like compounds. The observed regioselectiv-
ity is due to the annulation of the aromatic system to
an alicyclic ring, rather than due to mere substitution
at C-4a and O-1. The effect of the annulation strain on
the oxidation selectivity, that is, the ratio of the two pos-
sible ortho-quinone methides, can be smoothly explained
8. 4-Methoxy-2,3,6-trimethyl-5-(3-methyl-butyl)phenol (4),
1
white powder, mp 131–132 ꢁC. H NMR: d 1.07 (d, 6H,
CH–(CH3)2, 3J = 7.0 Hz), 1.60 (q, 2H, CH–CH2), 1.71 (m,
1H, CH(CH3)2), 2.05, 2.06, 2.12 (3 · s, 3 · 3H, 3 · CH3),
3
2.58 (t, 2H, ArC–CH2, J = 7.3 Hz), 6.12 (s, 1H, OH). 13C
NMR: d 12.1, 12.1, 12.8 (3 · CH3), 20.4 (ArC–CH2), 21.0
(CH–(CH3)2), 27.1 (CH–(CH3)2), 44.4 (CH–CH2), 56.9
(OCH3), 115.4, 117.6, 119.3, 121.1, 145.9, 148.2 (ArC).
4
by the SIBL modelas mentioned above.

7848
T. Rosenau, A. Stanger / Tetrahedron Letters 46 (2005) 7845–7848
Anal. Calcd for C15H24O2 (236.35): C, 76.23; H, 10.24.
Found: C, 75.99; H, 10.20.
Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.;
Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.;
Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.;
Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.;
Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P.
M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez,
C.; and Pople, J. A.; Gaussian, Inc., Wallingford CT,
2004; (b) All the systems under study underwent analytical
frequencies calculations to ensure minima (Nimag = 0) and
for obtaining zero point energies (ZPE) and thermochem-
icalparameters.
9. (a) Gaussian 03, Revisions B.05 and C.02, Frisch, M. J.;
Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M.
A.; Cheeseman, J. R.; Montgomery, Jr., J. A.; Vreven, T.;
Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.;
Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.; Scal-
mani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.;
Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa,
J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai,
H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.;
Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.;
Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A.
J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.;
Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J.
J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain,
M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.;
10. (a) See Ref. 2; (b) Rosenau, T.; Potthast, A.; Elder, T.;
Kosma, P. Org. Lett. 2002, 4, 4285–4286; (c) Rosenau, T.;
Habicher, W. D. Tetrahedron 1995, 51, 7919–7926.
11. If the small deviation from the 50/50 ratio is seen as a faint
indication of regioselectivity, then it is opposite to the
postulated effect.